Contents lists available at ScienceDirect
Applied Surface Science
journal homepage: www.elsevier.com/locate/apsusc
Towards Oxide Electronics: a Roadmap
M. Coll a , J. Fontcuberta a , M. Althammer b,c , M. Bibes d , H. Boschker e , A. Calleja f , G. Cheng g,h,i , M. Cuoco j , R. Dittmann k , B. Dkhil l , I. El Baggari m , M. Fanciulli n , I. Fina a , E. Fortunato p,q , C. Frontera a , S. Fujita r , V. Garcia d , S.T.B. Goennenwein s,t , C.-G. Granqvist u , J. Grollier d ,
R. Gross b,c,v , A. Hagfeldt w , G. Herranz a , K. Hono x , E. Houwman y , M. Huijben y , A. Kalaboukhov z , D.J. Keeble aa , G. Koster y , L.F. Kourkoutis ab,ac , J. Levy h,i , M. Lira-Cantu ad ,
J.L. MacManus-Driscoll ae , Jochen Mannhart e , R. Martins n,o , S. Menzel i , T. Mikolajick af,ag , M. Napari ae , M.D. Nguyen y , G. Niklasson u , C. Paillard ah , S. Panigrahi p,q , G. Rijnders y , F. Sánchez a , P. Sanchis ai , S. Sanna aj , D.G. Schlom ak,al , U. Schroeder af , K.M. Shen ab,ak , A. Siemon am ,
M. Spreitzer an , H. Sukegawa x , R. Tamayo f , J. van den Brink ao , N. Pryds aj , F. Miletto Granozio ap, ⁎
a
Institut de Ciència de Materials de Barcelona (ICMAB-CSIC), Campus of UAB, 08193 Cerdanyola del Vallès, Catalonia, Spain
b
Walther-Meißner-Institut, Bayerische Akademie der Wissenschaften, 85748 Garching, Germany
c
Physik-Department, Technische Universität München, 85748 Garching, Germany
d
Unité Mixte de Physique, CNRS, Thales, Univ. Paris-Sud, Université Paris-Saclay, 91767 Palaiseau, France
e
Max Planck Institute for Solid State Research, Heisenbergstr. 1, 70569 Stuttgart, Germany
f
OXOLUTIA S.L, Avda. Castell de Barberà, 26, Tallers 13, Nau 1, 08210 Barberà del Vallès, Barcelona, Spain
g
CAS Key Laboratory of Microscale Magnetic Resonance and Department of Modern Physics, University of Science and Technology of China, Hefei 230026, China
h
Department of Physics and Astronomy, University of Pittsburgh, Pittsburgh, PA 15260, USA
i
Pittsburgh Quantum Institute, Pittsburgh, PA 15260, USA
j
CNR-SPIN and Dipartimento di Fisica ''E. R. Caianiello", Università di Salerno, IT-84084 Fisciano (SA), Italy
k
Peter Grünberg Institut (PGI-7), Forschungszentrum Jülich GmbH, 52425 Jülich, Germany
l
Laboratoire Structures, Propriétés et Modélisation des Solides, CentraleSupélec,CNRS-UMR 8580, Université Paris-Saclay, 91190 Gif-sur-Yvette, France
m
Department of Physics, Cornell University, Ithaca, NY 14853, USA
n
Department of Materials Science, University of Milano Bicocca, Milano, Italy
o
MDM Laboratory, IMM-CNR, Agrate Brianza, Italy
p
CENIMAT/i3N, Departamento de Ciência dos Materiais, Faculdade de Ciências e Tecnologia (FCT), Universidade NOVA de Lisboa (UNL), Portugal
q
CEMOP/UNINOVA, 2829-516 Caparica, Portugal
r
Kyoto University, Katsura, Kyoto 615-8520, Japan
s
Institut für Festkörperphysik, Technische Universität Dresden, 01062 Dresden, Germany
t
Center for Transport and Devices of Emergent Materials, Technische Universität Dresden, 01062 Dresden, Germany
u
Department of Engineering Sciences, The Ångström Laboratory, Upp sala University, P.O. Box 534, SE 75121 Uppsala, Sweden
v
Nanosystems Initiative Munich (NIM), 80799 München, Germany
w
Laboratory of Photomolecular Science, Institute of Chemical Sciencesand Engineering, Ecole Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland
x
Research Center for Magnetic and Spintronic Materials, National Institute for Materials Science (NIMS), 1-2-1 Sengen, Tsukuba 3050047, Japan
y
MESA+ Institute for Nanotechnology, University of Twente, 7500 AE Enschede, Netherlands
z
Department of Microtechnology and Nanoscience - MC2, Chalmers University ofTechnology, Göteborg, Sweden
aa
Carnegie Laboratory of Physics, SUPA, School of Science and Engineering, University of Dundee, Dundee DD1 4HN, UK
ab
Kavli Institute at Cornell for Nanoscale Science, Ithaca, NY 14853, USA
ac
School of Applied and Engineering Physics, Cornell University, Ithaca, NY 14853, USA
ad
Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and The Barcelona Institute of Science and Technology (BIST), Campus UAB, Bellaterra, E-08193 Barcelona, Spain
ae
Department of Materials Science and Metallurgy, University of Cambridge, 27 Charles Babbage Road, Cambridge CB3 0FS, UK
af
NaMLab gGmbH, Noethnitzer Straße 64, 01187 Dresden, Germany
ag
Chair of Nanoelectronic Materials, TU Dresden, 01062 Dresden, Germany
ah
Physics Department, University of Arkansas, Fayetteville, AR 72701, USA
ai
Nanophotonics TechnologyCenter, Universitat Politècnica de València, Camino de Vera s/n, 46022 Valencia, Spain
aj
Department of Energy Storage and Conversion, Technical University of Denmark, DK-4000 Roskilde, Denmark
ak
Laboratory of Atomic and Solid State Physics, Department of Physics, Cornell University, Ithaca, NY 14853, USA
al
Department of Material Science and Engineering, Cornell University, Ithaca, NY 14853, USA
am
Institut für Werkstoffe der Elektrotechnik (IWE 2), RWTH Aachen University, 52066 Aachen, Germany
https://doi.org/10.1016/j.apsusc.2019.03.312
⁎
Corresponding author.
E-mail address: fabio.miletto@spin.cnr.it (F. Miletto Granozio).
0169-4332/ © 2019 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
T
an
Advanced Materials Department, Jožef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
ao
Institute for Theoretical Solid State Physics, IFW-Dresden, Helm- holtzstr. 20, D-01069 Dresden, Germany
ap
CNR-SPIN, Naples Unit, Complesso Universitario di Monte Sant’An -gelo, Via Cinthia, IT-80126 Napoli, Italy
Abstract
At the end of a rush lasting over half a century, in which CMOS technology has been experiencing a constant and breathtaking increase of device speed and density, Moore’s law is approaching the insurmountable barrier given by the ultimate atomic nature of matter. A major challenge for 21st century scientists is finding novel strategies, concepts and materials for replacing silicon-based CMOS semiconductor technologies and guaranteeing a continued and steady technological progress in next decades. Among the materials classes candidate to contribute to this momentous challenge, oxide films and heterostructures are a particularly appealing hunting ground. The vastity, intended in pure chemical terms, of this class of com- pounds, the complexity of their correlated behaviour, and the wealth of functional properties they display, has already made these systems the subject of choice, worldwide, of a strongly networked, dynamic and interdisciplinary research community.
Oxide science and technology has been the target of a wide four-year project, named Towards Oxide-Based Electronics (TO-BE), that has been recently running in Europe and has involved as participants several hundred scientists from 29 EU countries. In this review and perspective paper, published as a final deliverable of the TO-BE Action, the opportunities of oxides as future electronic materials for Information and Communication Technologies ICT and Energy are discussed. The paper is organized as a set of contributions, all selected and ordered as individual building blocks of a wider general scheme. After a brief preface by the editors and an introductory contribution, two sections follow. The first is mainly devoted to providing a perspective on the latest theoretical and experimental methods that are employed to investigate oxides and to produce oxide-based films, heterostructures and devices. In the second, all contributions are dedicated to different specific fields of applications of oxide thin films and heterostructures, in sectors as data storage and computing, optics and plasmonics, magnonics, energy conversion and harvesting, and power electronics.
Index Preface
M. Coll, J. Fontcuberta, N. Pryds and F. Miletto Granozio Introduction
1. A Unique Exploration J. Mannhart and H. Boschker Methods
2. What theoretical approaches can provide: a perspective on oxide electronics
M. Cuoco and J. van den Brink
3. Perspectives for applications of ultimate (atomic) control of oxide films using PLD
G. Koster, M. Huijben, G. Rijnders
4. Oxide MBE and the path to creating and comprehending artificial quantum materials
D. G. Schlom and K.M. Shen
5. Nanoscale patterning of complex-oxide materials A. Kalaboukhov and H. Boschker
6. Epitaxial oxide films on semiconductor substrates F. Sánchez and M. Spreitzer
7. Recent achievements and challenges in atomic layer de- position of complex oxides for heterostructures
M. Napari and J.L. Macmanus-Driscoll
8. Structure solving and refining, and strain gradients mapping in epitaxial thin films by X-ray diffraction techniques C. Frontera
9. Characterization of point defects in functional oxide thin films
D.J. Keeble
10. Developments in electron microscopy of exotic states at oxide interfaces: cryogenic imaging and advanced detec- I. El Baggari and L.F. Kourkoutis tors
Applications
11. Resistive switching oxides for data storage R. Dittmann
12. Oxides for data storage and processing: Ferroelectric tunnel junctions
V. Garcia and M. Bibes
13. Oxides for data storage: Ferroelectric RAMs U. Schröeder and T. Mikolajick
14. Alternative logic concepts using oxide-based electronic devices
S. Menzel and A. Siemon
15. High-k dielectrics for CMOS and emerging logic devices M. Fanciulli
16. Oxide nano-electronics for neuromorphic computing J. Grollier
17. Possible future quantum technologies based on correlated nanoelectronics
G. Cheng and J. Levy
18. Epitaxial oxide barriers for magnetic tunnel junctions H. Sukegawa and K. Hono
19. Magnetically ordered insulators for advanced spintronics M. Althammer, S.T.B. Goennenwein and R. Gross
20. Functional oxides in photonic integrated devices G. Herranz and P. Sanchis
21. Recent concepts and future opportunities for oxides in solar cells
A. Hagfeldt and M. Lira-Cantu 22. All-oxide heterojunction solar cells
R. Tamayo and A. Calleja 23. Photoferroelectrics
I. Fina, C. Paillard and B. Dkhil
24. Progress of indium-free transparent conducting oxides S. Panigrahi, R. Martins and E. Fortunato
25. Electrochromic and thermochromic oxide materials G.A. Niklasson and C.G. Granqvist
26. Ionotronics and nanoionics in energy devices: current status and future of μ-SOFC
S. Sanna and N. Pryds
27. Piezo-MEMS for energy harvesting
M.D. Nguyen, E. Houwman, G. Koster, and G. Rijnders 28. Gallium oxide for power electronics
S. Fujita
Preface
Mariona Coll, Josep Fontcuberta, Nini Pryds, Fabio Miletto Granozio In early 2007, motivated by the discovery of high mobility and quantum Hall effect in a transition metal oxide [1] A.P. Ramirez, en- thusiastically argued that ''the era of oxide electronics" had been finally entered [2]. Trying to summarize in a preface, more than ten years later, what makes oxide-based electronics potentially so special is still an arduous task. In the realm of oxides diversity and complexity play a key role. Oxide physics is not easily reduced to a few words.
The unparalleled wealth of oxide's functional properties is first rooted in the extreme diversity characterizing, in purely chemical terms, this materials platform. The introductory paper ''A unique ex- ploration" sharply clarifies that, at the closing of a technological era prevalently dominated by Si, oxides open the whole periodic table as the playground of tomorrow's materials scientists. Still, this spectacular chemical diversity only partially contributes to making oxides arguably the richest class of electronic materials. The sensitivity of oxides elec- tronic systems to their structural background is in fact such, that competing ground states with highly different properties can be also obtained under minimal (percental) chemical variations. Even at fixed composition, phase transitions can be induced under the effect of dif- ferent structural knobs modifying the systems boundary conditions: e.g.
thickness, strain, grain boundaries and interfaces.
The main reason that oxide materials are ''complex" (using this word in its literal physical sense [3]) can be largely traced back to the nar- rowness of their bandwidths. The chemical origin of this lies, in turn, in the ionic character of oxygen bonds, if compared with the highly covalent IV and III-V semiconductors. Bandwidth reduction becomes extreme in the case of transition metal (TM) oxides, dominated by the scarcely overlapping and partially occupied md
norbitals, with m=3, 4 or 5 and n = 1,... 10. For such materials the bandwidth W approaches the eV or even sub-eV energy scale. In this regime, a number of intrinsic degrees of freedom, e.g. electrostatic on-site repulsion, spin-orbit cou- pling, magnetic interactions and electron-lattice interactions, having comparable associated energies, come strongly into play in determining the system's electronic ground state. It's no surprise, therefore, that based on this intrinsic competition of charge, spin orbital and lattice degrees of freedom, TM oxides present the richest variety of emergent states and that new paradigms are needed to tackle their theoretical understanding, as explained in the second introductory contribution (''What theoretical approaches can provide: a perspective on oxides elec- tronics"), to both interpret the experimental data and foresee the properties of not-yet-synthesized materials.
The level of complexity rises even further when interfaces between oxides are considered. Adjacent materials affect each other through, e.g., charge transfer related to band mismatch, built-in fields, di- electric/ferroelectric surface charges, strain, structural distortions and magnetic (exchange, superexchange, Dzyaloshinskii-Moriya) interac- tions. The discovery of the existence of a high mobility and super- conducting electron gas at interfaces between large band gap insulating oxides [4] is a clear example of the new ''opportunities that interfaces bring to Oxide Electronics" [5]. In a new twist, emerging topological properties arising at interfaces and surfaces of oxide thin films are now driving a renewed attention [6]. In 2016, M. Lorentz and M.S. Ra- machandra Rao brilliantly brought together views from experts in the field and published an Oxide electronic materials and oxide interfaces road-map" that constitutes a snapshot of the exciting oxide-based sci- ence [7].
The maturity of the field, both in terms of materials, concepts and methods, now demands a renewed attention to address a few crucial questions about the potential technological and social benefits of a
research that has involved enormous amount of human, financial and technical resources in last decades. The purpose of the current paper ''Towards Oxide Electronics: a Roadmap", published as Final Action Dissemination of the TO-BE COST Action MP1308 (''Towards oxide- based electronics"), is to shift the focus, with respect to the previously mentioned effort [7], addressing more directly some crucial questions:
• What is the future role of oxide thin film in modern technologies?
• How far are oxides from taking this role?
• Which are the competing technologies?
• Which are the hurdles presently preventing the diffusion of mar- ketable applications?
• Which are the chances these hurdles can be overcome?
• Which are the advancement in synthesis and characterization methods that can help us facing them?
In making our selection of topics and in defining the index of the present paper, some filters were obviously applied. On the one hand, we tended to privilege the technologies and the ideas that already gained some recognition among the principal figures (companies and academic institutions) of the ongoing technological rush, with respect to others that remained so far mostly confined to the ''oxide" community. On the other hand, we focused explicitly on thin-film-based applications, and when possible on epitaxial films, deliberately neglecting vast sectors of oxide technologies based e.g. on bulk devices and on nanocrystalline samples/surfaces. Due to the latter choice, not only many well estab- lished technologies resorting to oxide components were ignored, but even some novel and very active fields of oxide research, as gas sensing and catalysis, were not considered.
Most of the applications presented in the following are expected to impact the fields of ICT and Energy. In times in which mankind is facing the extreme ambition of artificially emulating the computing power of a human brain, exceeding the exaflop range and corresponding, with the present energy-hungry electronic technologies, to a power consumption in the several GW range, the two fields are related as never before.
Selected topics of this Roadmap include: (a) Insights from theory and modelling, (b) advanced growth, nanofabrication and character- ization techniques and (c) applications on: data storage and computing, optics and plasmonics, magnonics, energy conversion and harvesting, and power electronics.
Our hope is that this Roadmap will show to be valuable both as a collective self-analysis of the oxide electronics community, providing an updated picture of the state-of-the-art in the field, and as a dis- semination tool, helpful to whoever is willing to spread knowledge about oxide science and to attract towards this exciting field of research new young scientists and further public and private resources.
Acknowledgments
This paper is based upon work from COST Action MP1308 "Towards
oxide-based electronics" (TO-BE), supported by COST (European
Cooperation in Science and Technology) - www.cost.eu. Josep
Fonctuberta and Mariona Coll express their acknowledgement for the
financial support from the Spanish Government, through the ''Severo
Ochoa" Programme for Centres of Excellence in R&D (SEV-2015-0496)
and the MAT2017-85232-R, and MAT2017-83169-R projects and from
Generalitat de Catalunya (2017 SGR 1377) during the preparation of
this document. M.C. acknowledges the Ramon y Cajal RyC2013-12448
contract. Nini Pryds is gratefully acknowledge the Danish Council for
Independent Research Technology and Production Sciences for the DFF-
Research Project 2 grants No. 6111-00145B (NICE) and the European
Union's Horizon 2020 research and innovation programme under grant
agreement No. 801267 (BioWings)
A unique exploration J. Mannhart *, H. Boschker
Max Planck Institute for Solid State Research, Heisenbergstr. 1, 70569 Stuttgart, Germany
* Corresponding author.
E-mail: office-mannhart@fkf.mpg.de (J. Mannhart).
Abstract
The Roadmap of Oxide Technologies for Electronic Applications provides a view into the future of oxide electronics. In this introduction to the roadmap, we scrutinize possible merits of oxide electronics, oxide electronics research, and the objectives of the roadmap.
Keywords: Oxides; Oxide electronics; Periodic table
Why is research valuable to our society? Which projects should be pursued? And, from the perspective of us scientists, what research is worth our own personal time and effort? Apparently, there are two obvious, diverging answers:
First, there is fundamental research to understand how the universe works. Such explorations are an essential trait of humans, which is why numerous space missions, and research institutions such as CERN and LIGO have received and continue to enjoy funding. It is less widely known, however, that profound questions about the basic properties of nature are also being answered by solid-state research. For example, current research seeks to resolve the wonders of macroscopic quantum states that raise quantum physics to the human scale, to reveal the existence and properties of novel quasi-particles, or to illuminate the astounding capabilities of emerging physical behaviors. It is worth noting that the study of electron systems in oxides is one of the methods with which these intriguing questions are being explored. For instance, consider the studies of the d-wave symmetry of the cuprate super- conductors' macroscopic order parameter (comparable to a gigantic molecular state) and the symmetry's influence on the current-carrying properties of high-T
ccables [8]. Or take the following, less applied examples: the search for a possible electric dipole moment of the electron using multiferroics to investigate possible charge-parity (CP) violations [9,10], the search for magnetic monopoles in spin ice [11,12], or the exploration of artificial atoms using complex oxides [13].
Second, research has sometimes led to applications that have im- proved our lives and may even be essential for our future, and this clearly provides abundant motivation to pursue such work. In fact, the most important, revolutionary applications developed to date have ty- pically arisen from curiosity-driven investigations. Often, the sub- sequent applications had not even been imagined when the original research was performed, as exemplified by spectacular results from solid-state science. For example, Kamerlingh Onnes' drive to cool matter to 4.2 K laid the base for magnetic resonance imaging (MRI) with its enormous value in medicine, or Michael Faraday's studies of electrical conductivity in Ag
2S [14], which led to the discovery of semiconductors, or—to mention but one such pioneer— Ferdinand Braun's investigations of rectifying metal contacts to PbS or FeS crys- tals, which induced the invention of the diode, the solar cell, the transistor, the integrated circuit and the internet. It will be exciting to see the unexpected applications to which oxide electronics will lead in the future.
In addition to these breakthrough discoveries, numerous technical advances have occurred in a more gradual and—some say—more pre- dictable manner. Our roadmap will provide an overview of these de- velopments in the field of oxide electronics, ranging from advances in film growth and the characterization of heterostructures at the atomic level to ferroelectrics, oxide photovoltaics, solid-state fuel cells, neu- romorphic computing, oxide microelectromechanical systems (MEMS), and transparent conductors.
Note that, whereas this roadmap illuminates the progress of oxide technologies and electronics from the perspective of applications, the advances achieved by improving fabrication technologies, analytical tools and our understanding of oxide devices will feed back into fun- damental science. In addition, addressing and questioning the limits of technology, which this roadmap intends to trigger, will continue to inspire scientists to raise fundamental questions about our physical world.
Another consideration we deemed relevant in putting this road-map into perspective was expressed by Rolf Landauer in a notable paper [15]. Based on his long-time experience at IBM Research, he vividly stresses in this sobering publication that only a small fraction of the innovative devices envisaged, invented, or developed by scientists ever become successful products. Although this is obviously true, Landauer's warning in no way calls for us not to invent novel devices, at least if common-sense is used in considering their possible value. As we know, some inventions will be quite successful, and often it is simply im- possible to foresee which of the many innovations will become winners.
Indeed, oxides have already made a huge impact on electronics. Con- sider for example (Ba,Sr)TiO3-based capacitors, PZT-based transducers, Hf-based oxinitrides for high-k gate insulators, and indium tin oxide as transparent conductors for displays, smart windows and photovoltaics.
These considerations underscore that research in an applied and scientifically exciting field such as oxide electronics is relevant for both, applications as well as fundamental science. But there is even more to oxide electronics. Oxide electronics is a trailblazer in a unique ad- venture: the pioneering endeavor of opening the entire periodic table to applications in electronic devices. To appreciate the relevance of this effort, we should recall the astounding fact that only a few but ex- tensively exploited chemical elements have driven the phenomenal success of complementary metal-oxide-semiconductor devices based on silicon (Si-CMOS), the workhorse of standard semiconductor tech- nology: silicon and germanium are used as semiconductors, oxygen is added to make SiO
2, copper and aluminum serve as interconnects, hafnium and nitrogen for gate dielectrics, and gold for the contacts. In addition to these pervasive Si-CMOS elements, a much larger number of elements and compounds are relevant constituents of industrial semi- conducting devices. To name but a few examples besides Si and Ge, compounds such as GaAs, InSb, GaP, CdTe and ZnO are obviously im- portant semiconductors. Furthermore, numerous elements such as B, As, P and Ga are dopants of choice. And let's not forget materials such as Se, Cu2O, and CuS, which played important roles in semiconducting devices of the past.
It is remarkable that virtually all of these materials comprise mean- field electron systems only. In essence, mean-field systems define the arena of materials for semiconducting electronics, because semi- conducting devices rely almost exclusively on mean-field phenomena.
Band diagrams are an essential tool for semiconductor engineers. Recall Herbert Kroemer's notable statement, ''If, in discussing a semiconductor problem, you cannot draw an Energy Band Diagram, this shows that you don't know what you are talking about" [16]). Mean-field behavior pro- vides the basis for the rules of band-diagram design. One of the few exceptions to the mean-field behavior of standard semiconductors is the quantum-Hall effect where, by applying magnetic fields, we can induce correlations in mean-field systems [13].
Although a fairly large and growing number of elements besides silicon and numerous compounds thereof are used in semiconductor technology today, they are but a subset of the periodic table, and the number of compounds used in semiconductor applications pales in comparison to the full range of electronic materials at our disposal.
In a vigorous development that was much initiated by the discovery of high-T
csuperconductivity in the copper oxides, the scientific com- munity is now making rapid progress in learning how to grow films and heterostructures using almost any element of the periodic table that is not a noble gas, super-toxic, or radioactive (see Fig. 1).
This endeavor is not restricted to pure elements. Countless
compounds are being grown, with chalcogenides and pnictides at- tracting particular interest. Of these compounds, oxides are especially suitable for device applications. As oxygen is highly electronegative, and O
2–has a small ionic radius and carries a double charge, numerous thermodynamically stable phases of many oxides can be synthesized under practical conditions, including phases that comprise several sorts of cations. Given the size and bonding properties of oxygen ions, oxides can crystallize in a wide range of crystal structures. The tendency of O
2–to bond with metal-ion d orbitals to form tetrahedra and octahedra that can readily be distorted, the variability of oxygen occupancy and doping, and the enormous polarizability of oxygen ions provide the basis for a broad spectrum of electronic properties, often including band gaps in the electron-volt range, which is desirable for applications in electronic devices as described in this roadmap [17].
The scientific community is developing increasingly refined tools to design, grow and characterize these materials with atomic precision.
The accuracy and supercell size of DFT calculations are being extended to inform the design of heterostructures and devices. Handling smaller supercells, DMFT calculations allow correlation effects to be assessed.
Advanced film growth techniques based, for example, on molecular beam epitaxy (MBE), pulsed laser deposition (PLD), and atomic layer deposition (ALD) are now available, and analytical tools such as scan- ning transmission electron microscopy (STEM) and electron energy loss spectroscopy (EELS), resonant inelastic X-ray scattering (RIXS) and angle-resolved photoelectron spectroscopy (ARPES), transport mea- surements and tunneling spectroscopy are being used with great success for characterization. With recent advances in stoichiometry control, substrate termination, and reflection high-energy electron diffraction (RHEED), it is now possible to grow heterostructures of complex com- pounds of surprising quality. Although defect concentrations and mo- bilities of semiconducting heterostructures are distinctly superior to those of complex oxides, according to their STEM images and EELS analyses, in terms of quality, oxide heterostructures already seem at least comparable to today's most advanced III-V heterostructures, an achievement that was unthinkable twenty years ago (see, e.g., [18]).
While the exploration of the oxides is of special interest for devices and is at the core of our oxide electronics roadmap, oxide electronics research has to be understood as part of a more general trend toward using complex compounds. The advances, possibilities, and challenges discussed in this roadmap of designing, growing and using oxides are valid in principle also for nitrides, sulfides and many other compounds, and to some extent even for organic systems. The ongoing adventure is to explore and leverage virtually all elements of the periodic table and compounds thereof for practical applications as well as for fundamental science.
As more elements become available for use in devices, the number of possible compounds grows astronomically [19]. The fabrication of highly complex compounds is of course a generic goal of chemistry and materials science. It is already possible today to grow many of such materials as epitaxial films and heterostructures, and to pattern them into nanodevices. In fact, with our capability to grow heterostructures with atomic precision, the phase space of electronic systems is be- coming virtually unlimited.
In this development, not only the number of available compounds is growing, but, more importantly, also their range of functional proper- ties. The mean-field compounds discussed above are based almost ex- clusively on s and p-electron systems. Thanks to transition-metal oxides and felements, electronic correlations are becoming increasingly re- levant by providing completely new arenas for device applications, such as phase-transition transistors, negative capacitors and photo- voltaic cells based on correlation effects, all of which are com- plementing and adding to mean-field Si-CMOS. Nevertheless, despite these bright prospects for device applications, we should be mindful of the leap from device demonstrations to successful products, as de- scribed by Rolf Landauer.
There is yet another issue to ponder: Exploring the periodic table to discover and synthesize materials for electronic devices will be a one- off research adventure because there is of course only one periodic table. In their ground states, chemical elements just possess a limited set of s, p, d and forbitals and linear combinations thereof, with no addi- tional orbitals to come. The effects we will find, the discoveries we hope to make, and the issues we will uncover regarding their applications will be what they are, as there will never be a new set of chemical elements. The periodic table is the one stomping ground we have.
The finite number of building blocks available to compose new materials has unavoidable implications for the future of oxide electro- nics, even for condensed-matter science. In the long term, novel ma- terials will predominantly be sought and discovered in increasingly complex compounds, particularly in quaternary, quinary and senary systems. In addition, compounds will be tailored by nanostructuring, which opens the door to fabricating an immense variety of ''artificial atoms" of complex materials and heterostructures with novel ground states and functional properties [13]. Precision in sample growth and characterization will become increasingly relevant, and reproducibility, especially between different research groups, will become more chal- lenging. As materials get more complex, marginally stable and me- tastable phases will become more prevalent, and they will host a greater variety and density of defects. It will become increasingly difficult to find novel emerging or functional properties that occur at a sizable energy scale.
Today is an exceptionally good time to develop oxide electronics because we have just recently managed to add d and felectrons to heterostructures of increasing variety, complexity and quality. Now is the time for pioneering work, to discover and utilize novel and large effects, working with compounds of still moderate complexity. The novel possibilities afforded by these advances provide the topics of this roadmap.
The adventure of harvesting the periodic table for materials that may be useful for electronic devices, green applications and basic sci- ence clearly justifies the efforts we must undertake. We submit that this adventure will be valuable to our society and worthy of our attention, Fig. 1. Exploring the periodic table to discover and synthesize materials for
electronic devices is a one-off research adventure because there is, of course,
only one periodic table of the elements in our universe. In this rendition, we
have highlighted those elements that in our view are practical building blocks
for films and heterostructures to be used at room-temperature. Fairly im-
practical seem only the radioactive elements and the noble gases. Note that
some of the highlighted elements may be difficult to use because of their
toxicity. Background: Photo of the Wolf-Rayet star WR 22 and the Carina
Nebula (Credit: ESO).
effort, and time. Indeed, it is imperative that we make good, sensible use of the elements available to us.
Acknowledgements
The authors gratefully acknowledge helpful discussions with Dieter Fischer.
What theoretical approaches can provide. A perspective on oxide electronics
Mario Cuoco
a,b,*, Jeroen van den Brink
ca
CNR-SPIN, IT-84084 Fisciano SA, Italy
b
Dipartimento di Fisica ‘‘E. R. Caianiello”, Università di Salerno, IT-84084 Fisciano SA, Italy
c